Isabelle Cudlitz
commited on
Commit
·
d24fb43
1
Parent(s):
ad7f287
Upload model
Browse files- README.md +18 -0
- adapter_model.bin +1 -1
README.md
CHANGED
|
@@ -2916,4 +2916,22 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 2916 |
### Framework versions
|
| 2917 |
|
| 2918 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2919 |
- PEFT 0.6.2
|
|
|
|
| 2916 |
### Framework versions
|
| 2917 |
|
| 2918 |
|
| 2919 |
+
- PEFT 0.6.2
|
| 2920 |
+
## Training procedure
|
| 2921 |
+
|
| 2922 |
+
|
| 2923 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 2924 |
+
- load_in_8bit: True
|
| 2925 |
+
- load_in_4bit: False
|
| 2926 |
+
- llm_int8_threshold: 6.0
|
| 2927 |
+
- llm_int8_skip_modules: None
|
| 2928 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 2929 |
+
- llm_int8_has_fp16_weight: False
|
| 2930 |
+
- bnb_4bit_quant_type: fp4
|
| 2931 |
+
- bnb_4bit_use_double_quant: False
|
| 2932 |
+
- bnb_4bit_compute_dtype: float32
|
| 2933 |
+
|
| 2934 |
+
### Framework versions
|
| 2935 |
+
|
| 2936 |
+
|
| 2937 |
- PEFT 0.6.2
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 21020682
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7fa590bfb288a746d015809055f6bd9a3486064d783041e94b359fc9e06308c6
|
| 3 |
size 21020682
|